find-key-internal-sources

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The required workflow explicitly performs retrieval from internal_messaging and document_store connectors (e.g., channel threads/messages and wiki/docs body text) and then uses that retrieved evidence as LLM context, which can include outsider-authored free text (e.g., messages/comments or pages written by non-operating users).

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill mandates loading and running the preflight script from ../user-context/SKILL.md at runtime (invoked as $sales:user-context), which is a required external content fetch whose executed preflight output (sales_preflight) directly controls prompts and behavior.